I am currently reading an Excel file with 2500 rows and 35 columns. I update a bunch of those cells (about 500) and try to write them back to Excel again.
But for some reason, this does not work. The cell updater node jumps from 0 to 75% immediately and then stays there forever before (after 30 minutes or so) I try to cancel all running executions.
I know that it’s quite a lot of different cells (500) in a big matrix (~90,000 cells) to change but is that a performance thing or do I maybe have somewhere a wrong setting.
Maybe some KNIMErs can share their experience here.
does it crash KNIME or is the node just failing? In case of the latter could you please provide the error message or the log?
I just tested it myself with a bigger excel files than yours in terms of number of columns and rows but only simple string as contents and updating ~3000 cells. I only had problems when I reduced my memory (in the knime.ini to 2GB) letting the node fail with a heap space error.
Could you share a workflow to try to reproduce the issue?
I tried to reproduce the error but that’s not possible without sharing confidential data.
A simple file with a few 1,000 cells works fine, formulas in Excel stay, conditional formats do too…
But in the original file, it keeps on executing on the 75% level which I also can only stop by killing the KNIME process in Windows task manager. I also checked the knime.ini and it says xmax = 8GB (which I assume should be enough).
Interesting is that I also cannot stop the execution.
I finally found the problem after some in depth research. The XLSX file the workflow writes to contained a reporting sheet with lots of external links using direct links as well as GETPIVOTDATA formulas pointing to external pivot tables in other files.
Not KNIME’s fault (of course not ) so I will close this post now…
) so I will close this post now… 
